Many bone fractures show compromised regeneration-healing slowly (delayed union) or failing to heal (non-union). These injuries can be complicated by pain and loss of function, often requiring revision surgery, and are a significant cost to society. A robust vascular response is required for proper bone regeneration, thus a lack of vascularization is a primary risk factor for compromised bone healing. Therefore, methodologies that enhance vascularization will promote greater healing of compromised fractures. Thrombospondin 1 (TSP1) and its closest family member thrombospondin 2 (TSP2) are extracellular matrix proteins and potent endogenous inhibitors of angiogenesis that are over-expressed in non-healing wounds. TSP1 and TSP2 are highly expressed during the fibrovascular phase of compromised fracture healing and an absence of TSP2 results in enhanced vascularization and bone formation during ischemic fracture healing. A dominant mechanism for the influence of TSPs on revascularization in tissues is the interaction of these ligands with the CD47 receptor, a pleiotropic receptor that strongly influences ischemia-reperfusion. However, the significance of TSP-CD47 interactions in enhancing ischemic fracture healing has not been studied. We hypothesize that TSPs are endogenous inhibitors of ischemic fracture healing by binding to CD47. Blocking TSP-CD47 binding will promote fracture vascularization and enhance bone regeneration in compromised fracture healing. To address this hypothesis we propose three specific aims. First we will examine the contribution of TSP1 and TSP2 to ischemic fracture healing by studying TSP1-null, TSP2-null, and double-knockout mice. As well, we will add back TSP2 using adenovirus. In Aim 2, we will ask whether CD47 regulates fracture vascularization and bone regeneration in compromised fracture healing by binding to TSP. We will examine ischemic fracture healing in CD47-null mice, and then evaluate whether a TSP1-based CD47 activating peptide inhibits ischemic fracture revascularization. Finally, we will ask whether disruption of TSP-CD47 ligand-receptor interaction following fracture healing using CD47 blocking agents, delivered either locally or systemically, can enhances fracture vascularization and bone regeneration in compromised healing. Successful completion of these studies could have significant near-term clinical implications for the patient with a compromised fracture.